Liver fibrosis (LF) is a pathological reaction of abnormal deposition of extracellular matrix (ECM), which is a key step of chronic liver disease is cirrhosis or even liver cancer. . my country is a large country of liver disease, hepatitis B virus, non-alcoholic fatty liver disease, etc. is common pathogenic factors in liver fibrosis. Studies have shown that timely diagnosis and clinical intervention can shorten the process of liver fibrosis, and early liver fibrosis can even reverse  . The “gold standard” of hepatic fibrosis is a puncture biopsy, due to its honesty and repetitiveness, it is difficult to monitor the monitoring means of monitoring changes in liver fibrosis . Non-invasive diagnostic criteria of hepatic fibrosis becomes a hot spot in the current research.
This study retrospectively collects various liver diseases require surgical resection of serological examination results, calculates four free serum diagnostic models, and evaluates four models for different fibrosis.
2. Materials and Methods
2.1. General Information
Retrospectively Collection of Qingdao University Affiliated Hospital June 2019 Liver Obstruction After December 2019, accurate liver fiber 272 patients with pathological installments. Among them, 197 men, an average age of 53.1 ± 13.3; 75 females, an average age of 47.3 ± 14.8. All patient serological examination results were obtained before surgery and were taken. This study has been licensed by the Medical Ethics Committee of the Affiliated Hospital of Qingdao University.
2.2. Serum Diagnostic Model Calculation
Aspartte Aminotransferase to Platelet Ratio Index, Apri γ-Glutamyltranspeptidase to Platelet Ratio, GPR), Fibrosis Index Based On The Four Factors, FIB-4) and AST / Alt ratio (AST / Alt)Ratio, AAR) The calculation formula is as follows:
AST U L ) / ( U /  APRI = AST / ULN
1 00 / PLT 1 / ) FIB-4 = years ) AST ( / (
1 0 9 / / L ) / U ) 1 /  (3) GPR = ( / ULN ) × PLT ( 9 / L  2.3. Hepatic fibrosis stage according to Chinese Medical Association  Liver fibrosis is divided into five issues: ≥ S1 is defined as liver fibrosis, ≥S2 is significant fibrosis, ≥S3 is progressive liver fibrosis, ≥S4 is cirrhosis, specific staging standard See Table 1 . Table 1 . Staging criteria of liver fibrosis
Table 1 . Hepatic fibrosis stages Statistical analysis Calculate patient clinical and serological characteristics using SPSS23.0. Use independent sample T test comparison measurement, χ 2 Check the comparison count data, compare the use of single factors ANOVA test, use the ROC curve and the curve to evaluate the diagnostic ability of serum diagnosis model, Delong test detection ROC curve. P The patient’s clinical parameters are statistically significant, and the age is statistically significant. Serological indicators are also meaningful during five divisions, and patients have clinical serological characteristics [ Table 2 . Four serological diagnostic models based on clinical and serological indicators Apri, GPR, FIB-4, and AAR identification of different packet liver fibrosis, such as Table 3 , APRI, GPR, FIB-4 The difference between the four groups is meaningful, and the difference between AAR in any group is not statistically significant. Table 2 . Clinical sErological Characteristics Of Different Stages Of Hepatic Fibrosis . Clinical serological characteristics of different liver fibrosis Table 3 . Serum Models of Different Groups of Liver Fibrosis Table 3 . Different liver Serum model for different packets of liver fibrosis, we use ROC curves to analyze the diagnostic effects of APRI, GPR, FIB-4 serum diagnosis model. For ≥S1, ≥S2, ≥S3 and ≥S4 four groups, APRI, GPR, FIB-4 have better diagnostic efficacy, where FIB-4 performance is optimal, AUC ranges (0.811 ~ 0.960), for ≥ The S1 period is the best diagnosis, AUC reaches 0.960, and the critical value is 0.850 in the maximum value (0.814), which is shown in FIB-4 Table 4 . The ROC curve of the serum model of different packets is shown in FIG. 1 , the Delong checkp is less than 0.05. Table 4 . ROC CURVE Analysis of Serum Models in Different Groups of Liver Fibrosis . ROC curve analysis of serum model of different liver fibrosis 4. Discussion Hepatic fibrosis is the liver Due to the pathological repair reactions of excessive deposition of extracellular matrices in various pathogenic factors, there are many disease factors.Diverse  . Common for chronic hepatitis virus infection, non-alcoholic fatty liver disease, alcoholic liver disease, etc., liver fibrosis is a key step in the progression of chronic liver disease is cirrhosis  . At present, the diagnosis of “gold standard” of liver fibrosis is the liver puncture biopsy, but there is a complication after puncture surgery and the price is expensive, and it is difficult to carry out multiple times. Therefore, clinical urgent needs of a non-invasive, repeatable liver fibrosis assessment method  .
Serum markers commonly used in hepatic fibrosis are divided into direct markers and indirect markers  . Direct markers include hyaluronic acid, HA), laminin, LNs, and study show that the direct markers of liver fibrosis are positively correlated with the level of liver fibrosis . Indirect markers such as platelet counts, aspartate amino transferase, alanine amino transferase, etc., can reflect liver cell damage, indirect reaction liver fibrosis. However, the above serum marker indicators are affected by their own metabolism. When other organ injuries can cause the level of these markers, the limits of the liver are low  .
Non-invasive diagnostic models calculated by single blood cleaning index, such as APRI, FIB-4, GPR, AAR, S index, and the like can increase fibrotic diagnosis accuracy  . This study retrospectively analyzes the patients with liver disease in our hospital. After surgery, hepatic fibrosis was obtained, and the hepatic fibrosis stage was more accurate compared to the puncture patient. Based on the commonly used serological indicators of our hospital, APRI, FIB-4, GPR, AAR four serum diagnostic models, T test proves APRI, FIB-4, GPR three models to ≥S1, ≥S2, ≥S3 and ≥S4 four Various groups of liver fibrosis are meaningful (P-value range 0.001 ~ 0.049) AAR has no statistical significance for different groups of liver fibrosis. In the three models, the FIB-4 index has the best diagnostic performance, AUC ranges (0.811 ~ 0.960), diagnosed ≥ S1 liver fibrosis, i.e., the performance of liver fibrosis, AUC is 0.960 (95% CI: 0.936 ~ 0.985), in line with the literature . The FIB-4 index combines three single indicators of AST, ALT, GGT. Sonnevel  has shown that FIB-4 index diagnosis excludes the sensitivity of hepatitis B hepatic hepatic cirrhosis over 30 years of aged 90.9%. This study proves that APRI, FIB-4, and GPR three serum models can suggest that hepatic fibrosis, wherein FIB-4 diagnosis is optimal, and when FIB-4 has certain limitations, this study The research sample is limited, and the amount of liver fibrosis in different installments is uneven. S1, S3 period is small, which may result in packet errors. It is desirable to increase the sample amount, improve measurement standards, and the quantization value of serum model is further Accurately predict the liver fibrosis of different stages. Notes
* Corresponding author.
Reference < 0.05表示差异有统计学意义。
Sun, M. And Kisseleva, T. (2015) Reversibility of Liver Fibrosis. Clinics and Research in Hepatology and Gastroenterology, 39 , S60-S63. https://doi.org/10.1016/j.Clinre.2015.06.015
Ding Deping, Liu Ping, Chen Linli, et al. Fibroscan, AAR, APRI, FIB-4 and its joint application diagnosis of hepatic fibrosis in patients with hepatitis B The meaning of value [J]. Chinese and Western Medicine Combination Liver Disease Journal, 2016, 26 (5): 267-269 + 286.
Zhang Xu, Wang Wei, Ma Juan, et al. Comparison of the diagnostic value of 5 non-invasive diagnostic techniques for chronic hepatitis B hepatic fibrosis [J]. Journal of Clinical Hepatobiliary Diseases,2016, 32 (10): 1888-1893.
Lu Lungen, especially red, Xie Weif, et al. Hepatic fibrosis Diagnosis and treatment consensus (2019) [J]. Clinical Hepatobiliary Journal, 2019, 6 (10): 793-803.
Qin Yucheng, Chen Shenghua. Progress in the role and mechanism of extracellular matrix on hepatic star cells in the process of liver fibrosis [J] .Journal of Practical Liver Disease, 2017, 20 (3): 381-384.
Friedman, SL AND Bansal, MB (2006) Reversal of Hepatic Fibrosis -FACT OR FANTASY? HEPATOLOGY, 43, 82-88. < 0.850时可以排除肝纤维化。其他模型的AUC及95% CI如 https://doi.org/10.1002/hep.20974  Zhang Qi Di, Lu Lungen. The current status of non-invasive liver fibrosis / cirrhosis evaluation technology [J] Chinese Journal of Hepatosis, 2018, 26 (5): 325-327.
 Stibbe, KJM, Verveer, C., Francke, J., et al. (2011) Comparison of Non-Invasive Assess to Diagnose Liver Fibrosis in Chronic Hepatitis B and C Patients. Scandinavian Journal of Gastroenterology, 46, 962-972.
Liu Ting, Xu Mingyi. Progress in serological diagnosis of hepatic fibrosis [J] .Journal of Practical Liver Disease, 2016, 19 (1) : 125-128.
Lemoine, M., Shimakawa, Y, Nayagam, S., et al. (2016) The gamma -Glutamyl transpeptidase to Platelet Ratio (GPR) Predicts Significant Liver Fibrosis and Cirrhosis in Patients with Chronic HBV Infection in West Africa. Gut, 65, 1369-1376. < 0.850时可以排除肝纤维化。一项研究认为 ，GPR比APRI和FIB-4更准确，可以用于肝纤维化分期。张旭  等研究表明联合APRI、FIB-4在内的五种血清模型诊断肝纤维化的性能与病理结果趋近。这些诊断模型大多为慢性乙型肝炎或慢性丙型肝炎所致肝纤维化样本，目前对每种模型的临界值尚未达成一致 。
Nakamura, Y., Fukuhara, T., Aikata, H., et al. (2017) Liver Fibrosis Assassment by Fibroscan Compared with pathology, 48, 767-772.